No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

Heavy Ping drop issue due to layer two loop

Publication Date:  2014-10-27 Views:  150 Downloads:  0
Issue Description
Hardware: E9000 Chassis,CH240 with MZ510 card,CX310,S6700
The customer are facing heavy ping drop within E9000 chassis. They have installed ESXi5.1 on CH240 (MZ510) Blade#1 to Blade#4. They can't reach the ESX5.1 server IP from client network,even they can't reach the IP from CX310 which is directly connected to each balde.
The IP address and vlan setting of vmnic0 is as below:



Alarm Information
Many ping packages droped from CX310 to the blade:
Handling Process
We ask the customer provide us detail description of topology and configurations for each CX310 switch module:

Each CH240 was equiped with one MZ510 card in Mezz1,two 10GE ports connected to 2x-CX310 and 3x-CX310 separately.
2x-CX310 and 3x-CX310 connected to two S6700 separately,two S6700 connected to CISCO S6509 core switch.
The vlan gateway IP 192.168.45.1 was configed on CISCO S6509.

The internal ports of 2x-CX310 10ge 2/1/1 to 2/4/1 connect to blade#1 to blade#4 vmnic0;
The internal ports of 3x-CX310 10ge 3/1/1 to 3/4/1 connect to blade#1 to blade#4 vmnic1;
Internal ports configuration are:

External ports of 2x-CX310 10ge 2/17/1 to 2/17/4 connect to S6700A as uplinks;
External ports of 3x-CX310 10ge 3/17/1 to 3/17/4 connect to S6700B as uplinks;
External ports configuration are:



Root Cause
The topology is as below:

Combined with the configration and the topology,we can know:
1: The customer misunderstand the CX310 as a pass-through module,he thought each internal port must map to an external port;
2: There are four uplinks from CX310 to S6700 for both 2x and 3x ,but the customer didn’t config link aggregation between them.
So there are two layer-2 loops between 2X-CX310 and S6700A,3X-CX310 and S6700B.
This is the root cause for ping drop issue.
Solution
1: Explain to the customer that Cx310 is just like a switch,not a pass-through module,so there is no fixed map relationship between internal ports and external ports;
2:  Config link aggregation between 2x-CX310 and S6700A,3x-CX310 and S6700B .
Below is an example configuration in 2x-CX310:
interface Eth-Trunk2
description Trunk-Between-E9000-CX310-X2 to S6700A Switch
port link-type trunk
port trunk allow-pass vlan 2 to 4094
trunkport 10ge 2/17/1 to 2/17/4
mode lacp


After link aggregation,the topology is as below,so there is no loop,the problem was resolved.
Suggestions
None.

END